ABSTRACT: Micro discrete fracture networks (µDFN) have been embedded into Grain Boundary Models (GBM) within UDEC to simulate laboratory tests on a series of progressively larger in size numerical specimens. The initial GBMs simulating the intact rock consist of Voronoi blocks to capture the fracturing of the intact material. The µDFNs were generated separately from the GBMs in SR2 to create pre-existing “defects”. These geometries were created stochastically with fracture orientation, mean length and areal intensity (P21) serving as input parameters for the generation process. Following the generation of the µDFN geometries, the fracture networks were incorporated into the GBMs to create synthetic rock mass (SRM) models. This type of model is able to capture the behaviour of “flawed” rock samples as the defects are explicitly modelled. A sensitivity analysis is undertaken to examine the effect of micro-defect intensity on the UCS and modulus of deformation under an assumption of a constant crack length. Preliminary results highlight the impact of pre-existing defects and their geometrical configuration on the rock block strength.